Line ending backslash in """ causes later line breaks to also be removed

[JoelColledge]

Hi, I'm having trouble with breaking long lines in """ delimited strings using a backslash. This is called a "line ending backslash" in the toml spec.

Reproducer:

package main

import (
	"fmt"
	"os"

	"github.com/BurntSushi/toml"
)

const in = `f = """
a\
b
c
"""`

func main() {
	var out map[string]string
	if _, err := toml.Decode(in, &out); err != nil {
		fmt.Fprintf(os.Stderr, "error: %v\n", err)
		os.Exit(1)
	}

	fmt.Println(out["f"])
}

Expected output, which is generated with BurntSushi/toml v0.3.1:

ab
c

Output with BurntSushi/toml v1.2.1:

abc

Any chance of fixing this? Thanks!

[arp242]

This probably broke quite a few releases ago, as this hasn't been touched since last year; I'm guessing with 35432de.

I should probably rewrite that entire logic; I'm having trouble following it now.

[cespare]

Yikes. I sent #391.

@arp242 would you mind giving that a review and then cutting a 1.2.2 release? Or I'm happy to do it, though I figure you would have a better idea of what all to mention in the release notes.

[arp242]

would you mind giving that a review and then cutting a 1.2.2 release? Or I'm happy to do it, though I figure you would have a better idea of what all to mention in the release notes.

Let me take a look at #371 before the release; then every (known) bug should be fixed.

[cespare]

@arp242 I took at stab at that in #392 to try to speed things along.

However, I would say that this is a much more serious issue since (I'd guess) orders of magnitude more code depends on decoding than encoding. This bug caused a configuration issue at my work which is how it came to my attention.

[arp242]

I would say that this is a much more serious issue since (I'd guess) orders of magnitude more code depends on decoding than encoding.

I agree; I would (should) have fixed it much earlier; I just haven't been able to spend as much time on things in the last few months as I'd like due to some personal issues; I've been picking up things again since last week, but there's a bit of a backlog (so many emails...). I actually did create a quick fix yesterday, but your PR pre-empted that (and it's nicer anyway).

However, it seems to have been broken since June 2021, so it's less often used than I would expect, and in that context a few more days probably won't matter too much, I think? It avoids having to push out a new bugfix release just for that.

[cespare]

However, it seems to have been broken since June 2021, so it's less often used than I would expect,

One thing that makes that a bit tricky to evaluate is that people might not be pulling in the latest version of this package very quickly in well-established code bases with a lot of TOML given how MVS works. At my work we only pulled in this bug about five months ago.

and in that context a few more days probably won't matter too much, I think? It avoids having to push out a new bugfix release just for that.

Sure, a few days is fine by me. And as I said I'm also happy to run the release if that would help.

[arp242]

I tagged v1.3.0.

[cespare]

@arp242 thanks!